A predictive framework using advanced machine learning approaches for measuring and analyzing the impact of synthetic agrochemicals on human health

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: May 3, 2025

Pesticides and other synthetic agrochemicals play a critical role in emerging agricultural practices by enhancing crop productivity protecting against pests diseases. However, their widespread application has raised significant concerns about environmental balance adverse human health impacts, including neurological disorders, cancers, respiratory metabolic effects, particularly among workers vulnerable populations. Extensive literature underscored the detrimental consequences of pesticides on health. Although, incorporation machine learning algorithms for accurate risk evaluation predictive modeling still underexplored, requiring novel solutions. This study investigates impact using advanced techniques, leveraging multi-level feature selection, hybrid ensemble learning, SHAP, custom loss function to improve prediction accuracy. presents robust framework assessing risks posed agrochemicals, offering insights into assessment strategies. Data sourced from credible organizations, WHO, CDC, EPA, NHANES, USDA, underwent extensive preprocessing analysis. Machine (ML) models such as Random Forest, LightGBM, CatBoost were employed alongside selection methods like mutual information gain (MI) Recursive Feature Elimination (RFE). A is leveraged accurately predict mortality cases avoid misclassifications penalizing false negatives. Furthermore, Particle Swarm Optimization (PSO) Genetic Algorithm (GA) used model optimization. Results demonstrate superiority models, with LightGBM-PSO + CustomLoss achieving highest performance accuracy (98. 87%), precision (98.59%), recall (99.27%), F1 score (98.91%). Findings this can contribute policy making regulatory public safety Future directions will emphasize multi-regional dataset well external validation also real-world testing integration monitoring systems.

Language: Английский

---

Optimizing Constrained Engineering Optimization Problems Using Improved Mountain Gazelle Optimizer

Applied Computational Intelligence and Soft Computing, Journal Year: 2025, Volume and Issue: 2025(1)

Published: Jan. 1, 2025

This study introduces a novel approach to engineering design optimization through the development of an improved mountain gazelle optimizer (iMGO) that incorporates variable neighborhood search (VNS) techniques. The enhanced algorithm effectively addresses challenges by identifying optimal solutions within specified constraints. In particular, iMGO significantly improves solution diversity and mitigates risk premature convergence local optima, thereby overcoming limitations original MGO. A comprehensive analysis was conducted using 12 functions from CEC 2022 benchmark suite, applied five problems, including I‐beam, pressure vessel, three‐bar truss, cantilever beam, tension spring. Comparative results indicate outperforms established metaheuristic techniques, such as MFO, WOA, GOA, MPA, TSO, SCSO, well validate iMGO’s effectiveness in navigating complexities constrained optimization. For instance, practical applications, manufacturing cost vessel reduced 6014.4537 5915.3358, weight spring decreased 0.0149154 0.0130101 relative These enhancements underscore significant potential real‐world applications across aerospace engineering, structural optimization, energy system planning, other fields, contributing more efficient sustainable solutions.

Language: Английский

---

Climate change prediction in Saudi Arabia using a CNN GRU LSTM hybrid deep learning model in al Qassim region

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: May 10, 2025

Climate change, which causes long-term temperature and weather changes, threatens natural ecosystems cities. It has worldwide economic consequences. change trends up to 2050 are predicted using the hybrid model that consists of Convolutional Neural Network-Gated Recurrent Unit-Long Short-Term Memory (CNN-GRU-LSTM), a unique deep learning architecture. With focus on Al-Qassim Region, Saudi Arabia, assesses temperature, air dew point, visibility distance, atmospheric sea-level pressure. We used Synthetic Minority Over-sampling Technique for Regression with Gaussian Noise (SMOGN) reduce dataset imbalance. The CNN-GRU-LSTM was compared 5 classic regression models: DTR, RFR, ETR, BRR, K-Nearest Neighbors. Five main measures were evaluate performance: MSE, MAE, MedAE, RMSE, R². After Min-Max normalization, split into training (70%), validation (15%), testing (15%) sets. paper shows beats standard methods in all four climatic scenarios, R² values 99.62%, 99.15%, 99.71%, 99.60%. Deep predicts climate well can guide environmental policy urban development decisions.

Language: Английский

---

Optimized SVR with nature-inspired algorithms for environmental modelling of mycotoxins in food virtual-water samples

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: May 13, 2025

The accurate determination of mycotoxins in food samples is crucial to guarantee safety and minimize their toxic effects on human animal health. This study proposed the use a support vector regression (SVR) predictive model improved by two metaheuristic algorithms used for optimization namely, Harris Hawks Optimization (HHO) Particle Swarm (PSO) predict chromatographic retention time various mycotoxin groups. dataset was collected from secondary sources train validate SVR-HHO SVR-PSO models. performance models assessed via mean square error, correlation coefficient, Nash-Sutcliffe efficiency. outperformed existing methods 4-7% both learning (training testing) phases respectively. By using optimization, parameter adjustment became more effective, avoiding trapping local minima improving generalization. These results demonstrate how machine metaheuristics may be combined accurately forecast levels, providing useful tool regulatory compliance monitoring. framework perfect commercial quality assurance, testing, extensive programs because it provides exceptional accuracy resilience predicting times. In contrast conventional models, effectively manages intricate nonlinear interactions, guaranteeing identification while lowering hazards

Language: Английский

---

IoT Traffic Parameter Classification based on Optimized BPSO for Enabling Green Wireless Networks

Engineering Technology & Applied Science Research, Journal Year: 2024, Volume and Issue: 14(6), P. 18929 - 18934

Published: Dec. 2, 2024

The rapid expansion of artificial intelligence (AI) integrated with the Internet Things (IoT) has fueled development various smart devices, particularly for city applications. However, heterogeneity these devices necessitates a robust communication network capable maintaining consistent traffic flow. This paper employs Machine Learning (ML) models to classify continuously received parameters from diverse IoT identifying necessary adjustments enhance performance. Key parameters, such as packet data, are transmitted through gateways via specialized tools. Six different ML techniques default were used: Decision Tree (DT), Random Forest (RF), Support Vector Machines (SVMs), K-Nearest Neighbors (KNN), Naive Bayes (NB), and Stochastic Gradient Descent Classifiers (SGDC), environment (IoT / non IoT). models' performance was evaluated in real-time laboratory comprising 38 vendors following metrics: Accuracy, F1-score, Recall Precision. RF model achieved highest Accuracy 95.6%. Also Binary Particle Swarm Optimizer (BPSO) used across RF. results demonstrated that BPSO-RF hyperparameter optimization enhanced 95.6% 99.4%.

Language: Английский

---

Comparison of Convolutional Neural Network and MobileNetV2 for Potato Leaf Disease Detection

Published: Dec. 12, 2024

Language: Английский

---